Video wird geladen...
Video konnte nicht geladen werden
I was bored, so I created a self-learning Flappy Bird from scratch using a neural network and a genetic algorithm.
4,497,061 Aufrufe • vor 1 Jahr •via X (Twitter)
50 Kommentare

Artificial neural network:

Behind the scenes:

😂😂😭

Hope this helps:

Hope this help:

🥺 no banger badge

Me too Using ppo

Damn 🔥

Genetic algo is underrated 🤌

That's so true, fr.

Why no gen 8 showing thousands of successful dots?

Video length would have gone past 2 min.

no one: my brain: hey, remember the double-slit experiment?

Haha, I thought that too.

these ones decided the secret to life is cheating apparently

@LCChroma Haha lol.

Damn!

hehe. thanks.

@AkermanMartin Haha true.

1:25 he connected with the block mann.

@Domagoj_Gidija Agi 😭

Hard🗣️

Amazing! Can we implement this, but to catch a shiny Rayquaza in Pokémon Emerald?

Crazy Idea. I will ponder on this. Thanks bro❣️

I am willing to work on this together

Made one long ago as well, with the og NEAT Algorithm implemented from scratch myself (no libraries) Evolutionary algorithms really are underrated

I once made a bot to PK people in World of Warcraft. Watching it to see if it could beat a human player was more fun than playing the game normally...

Insane bro🔥😮💨

@haydendevs launch it on PlayStation

This is fire wtf

Post deemed a Standard banger.

Lets go❣️🔥

Cooool👌

I watched this for way too long. Viral TikTok material right here

this is what you'd be going against if u make a skill based game with a price tag @EbisuEthan :)

DOPE

Thank you.

Wow that was fast

The rate of improvement is impressive.

🙌

very cool

Is it possible to view the source code somewhere?

@grok tu m'explique ce que cette personne à fait ?

Cool

Nice, this is cool 🤜🤛

Thanks, bro.

It's only learning if you can randomize the gate height each playthrough. Otherwise it's just attrition.

love thisss🔥🔥

i wanted to ask how easy it is but i figured its not easy

Generation 7 lets go!
